The overall objective is to determine the hemopoietic potential of human and murine cells which give rise to colonies in diffusion chambers implanted intraperitoneally in mice (CFU-DG). By using chromosome markers and by utilizing subculture techniques, we wish to find out whether cells in CFU-DG colonies, which are derived from single cells, can, in addition to spleen colonies in lethally irradiated mice (12-day CFU-S), give rise to primitive blastic and mixed colonies in methylcellulose. We will determine whether CFU-DG can give rise to cells which will reconstitute the lymphoid system in lethally irradiated or severely immune-deficient mice. We also plan to determine whether cells from mixed colonies and blastic colonies can give rise to colonies in diffusion chambers in mice. These experiments should provide results which enable us to determine the hemopoietic potential of CFU-DG and its place in the hemopoietic cell hierarchy. This information is essential for interpretation of data, which involves effects on colony formation in diffusion chambers implanted intraperioneally following host treatment. Detection and partial purification of T- and B-cell-derived humoral activities which act on CFU-DG, but not on other colony-forming cells, further emphasize this need. Ideally, CFU-DG assay may be a probe for very early hemopoiesis.